Hi all, running into some weird behavior on tap-facebook. Seven of our eight facebook pipelines are running smoothly. One pipeline keeps timing out. Authorization is fine, since the tap connects to the Facebook servers, finds data, and completes to around 87% each time. Then the tap hits an unknown snag of some sort, causing it to initiate a series of back off protocols from which it never recovers. Log files have been checked, and yield no more detailed information. We’ve also tried re-running with “--log-level=debug' in search of more clues to how the error is happening, to no avail. Here's my config:

- name: tap-facebook
    variant: singer-io
    pip_url: git+<https://github.com/singer-io/tap-facebook.git>
    config:
      start_date: '2023-08-16'
      end_date: '2023-08-17'
      insights_buffer_days: 1
    select:
    - ads_insights_platform_and_device.*
    metadata:
      ads_insights_platform_and_device:
        replication-method: incremental
        replication-key: date_start
        date_start:
          is-replication-key: true

...and the error message:

CRITICAL Insights job 817604469899463 did not complete after 30 seconds. This is an intermittent error and may resolve itself on subsequent queries to the Facebook API. You should deselect fields from the schema that are not necessary, as that may help improve the reliability of the Facebook API